SNP genotype calling and quality control for multi-batch-based studies.

Area of Science:

• Genetics
• Bioinformatics
• Statistical Genetics

Background:

• Batch effects introduce systematic differences in genetic analyses due to batch heterogeneity.
• Effective quality control (QC) is crucial, but current statistical methods inadequately address batch effects.
• Novel approaches are needed for robust control of batch effects in genetic studies.

Purpose of the Study:

• To propose a new method for detecting probe intensity heterogeneity among batches.
• To develop a procedure for genotype calling and QC in the presence of batch effects.
• To improve the accuracy of genotype calling while minimizing data loss.

Main Methods:

• Multivariate analysis of variance (MANOVA) to test for probe intensity differences between batches.
• K-medoid clustering algorithm applied to probe intensity averages for subject stratification.
• Separate genotype calling for subjects within identified clusters.

Main Results:

• The method was applied to genotyping data from 3619 subjects, including Alzheimer's disease patients, mild cognitive impairment (MCI) patients, and controls.
• Improved accuracy in called genotypes was achieved.
• The approach effectively controls batch effects without requiring extensive filtering of subjects or single nucleotide polymorphisms (SNPs).

Conclusions:

• A novel strategy was developed to detect batch effects using probe intensity measurements and call genotypes accordingly.
• The proposed method provides a balanced approach for managing batch effects in real-world genetic data.
• The methodology is adaptable and can be extended to diverse genetic analysis scenarios with minor modifications.